An update about artificial mastication Marie-Agnès Peyron, Alain Woda

To cite this version:

Marie-Agnès Peyron, Alain Woda. An update about artificial mastication. Current Opinion in Food Science, Elsevier, 2016, 9, pp.21-28. <10.1016/j.cofs.2016.03.006>.

HAL Id: hal-01594489 https://hal.archives-ouvertes.fr/hal-01594489 Submitted on 26 Sep 2017

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés.

Available online at www.sciencedirect.com

ScienceDirect

An update about artificial mastication

1,2 3,4

Marie-Agne` s Peyron and Alain Woda

Developing masticatory apparatus, chewing robots or an for biomechanical studies. Secondly, in very different

artificial mouth is an old but ever more important goal in food approaches, simulators can be used to study either food

science, nutrition or dental research fields, as reflected by the bolus characteristics or to produce boluses for subsequent

number of existing digital or biomechanical systems. Whatever analyses (Figure 1).

the objective of the approach, basic knowledge of the

physiology of mastication, adaptation and neurophysiological Despite the important understanding gathered in sev-

control is absolutely needed before conceiving an apparatus. eral aspects of the masticatory process, simulation of

Obviously, the final step in the development of a mastication mastication in the area of food science has, too often,

simulator is its validation before performing food or food bolus been over-simplified and reduced to grinding, probably

characterization. This validation step is imperative to avoid due to the lack of knowledge of physiology. This

biased interpretation and can be performed through in vivo–in review resumes the main physiological key points of

vitro comparison of particle size distributions in food boluses masticatory process, and describes the different existing

obtained after normal mastication. This kind of validated simulations with biomechanical and modalities of func-

machine offers the chance to produce boluses for other related tioning.

uses such as nutrient bioaccessibility or digestion studies, for

example. Such an apparatus can also be employed to simulate Mastication must be understood before being

different dental states or ageing conditions. simulated

Through a complex and well-coordinated sensory-mo-

Addresses

1 tor and visceral activities, mastication of a solid mouth-

National Institute of Agronomic Research, Joint Research Unit 1019 for

Human Nutrition, Saint Gene` s Champanelle, France ful results in a bolus made of particles reduced in size,

2

Clermont University, University of Auvergne, Joint Research Unit

moistened enough to be cohesive, plastic to avoid

1019 for Human Nutrition, Clermont-Ferrand, France

3 particle aspiration, and to permit passage through the

Clermont University, University of Auvergne, CROC EA 4847,

throat without discomfort or pain. The sensory-motor

Clermont-Ferrand, France

4

CHU Clermont-Ferrand, Odontology Service, Clermont-Ferrand, and visceral program is continually commanded by the

France central nervous system. The food properties are sensed

as early as the first bite and, through sensory-motor

Corresponding author: Peyron, Marie-Agne` s

feedback, the masticatory program is adjusted to

([email protected])

the changes in bolus features occurring along the

masticatory process. This highly complex and feed-

Current Opinion in Food Science 2016, 9:21–28

back-dependent dynamic complicates any attempt to

This review comes from a themed issue on Sensory science and reproduce instrumentally mastication. Therefore, ad-

consumer perception vanced knowledge about how food structure influences

Edited by Susana Fiszman the pattern of oral processing is required. Food is a

complex stimulus, but the physical dimensions modu-

lating the oral processing are limited to its hardness, its

rough rheological dimensions (plasticity, elasticity or

http://dx.doi.org/10.1016/j.cofs.2016.03.006 brittle nature for example), and size of the mouthful.

2214-7993/# 2016 Elsevier Ltd. All rights reserved. Briefly, an increase in food hardness as well as in

mouthful size leads to an increase in the number of

masticatory cycles (tooth strokes) and applied muscle

forces, whatever the rheological nature of the food. On

the other hand, the rheological properties of food seem

mostly to impact the kinematics of mandibular move-

Introduction ments due to a need to adjust the combination of



Two main driving-objectives can be identified while compression and shear stresses [1 ]. Furthermore, frac-

simulating mastication: firstly, when the goal is to ture propagation during mastication inside the food

improve knowledge, to reproduce the biomechanical matrix strongly depends on its structure [2]. The num-

aspects of the masticatory system or to analyze the effect ber of fractures and consequently of food fragments

of forces, movements or constraints, for example. It seems mainly to depend on food toughness [3] with

generally results in the development and the use of resistant food often favouring fracture propagation,

mathematical models alone in an in silico approach or resulting in greater comminution. In parallel, the many

associated with mechatronic techniques to develop robots and well-documented individual chewing strategies

www.sciencedirect.com Current Opinion in Food Science 2016, 9:21–28

22 Sensory science and consumer perception

Figure 1

FOOD

controll ed

prog ramming parameters deficient programming

downg raded

masti cation simulator adjustement mastication simulator

param eters

mastication/ volunteers mastication /in vitro defi cient

masti cation /in vitro

in vi vo in vitro incorrect in vitro

foo d bolus food bolus foodbolus

analyses analysis / granulometry analysis / granulometry

in vivo in vitro in vitro / deficient mastication compliant ? no

yes

calib rated mastication simulator

validat ed food bolus (gold standard)

Various analyses:

dat a analysis Various analyses:

bolus characterisation bolus characterisation

kinetics of formation kinetic sof formation

rheology, granulometry rheology, granulometry

gold standard

saliva action, deficiency saliva action,

oral nutrient bioaccessibility oral nutrien t bioaccessibility

oral digestion deficiencies oral digest ion gold standard

bolus for GI digestion bolus for Gi digestion … …

data analysis

current opinion in food science

Flowchart displaying the key steps in development through sequential in vivo–in vitro actions, and validation stage of a mastication simulator

before operating it to produce boluses for multiple purposes.

1) Mastication of solid food ends with a bolus swallow-

help to accomplish the mechanical food disruption. The

able without risk of mucosal injury and aspiration. For

end point of the masticatory sequence is determined by

each food, a correct and specific granulometry,

the intrinsic properties of the bolus. Thus, swallowing

rheology and saliva impregnation characterize a

is initiated when the bolus has been perceived by the

swallowable bolus. In normal mastication, bolus

oral receptors to be ready for safe-swallowing. Thus the

particle size distribution is specific to food structure

swallowing threshold is a combination of numerous

and similar between boluses from different subjects.

physical dimensions including particle size, cohesive-

2) If such a bolus cannot be produced, mastication must

ness, elasticity, plasticity, moistening, intrinsic action of

be considered as impaired. At the individual level, two

mucines and enzymes, among other factors. In particu-

indicators sign for an impaired mastication: increased

lar, particles must be bound together by viscous forces

bolus granulometry above a certain threshold level and

rendering the bolus sufficiently cohesive [4,5]. This

variation in frequency of the strokes while masticating

swallowing threshold is specific to each food.

a given food compared with normal mastication.

3) In subjects with perfectly healthy mastication,

In summary, the basic points to be considered, before

increasing either the force or the number of tooth

simulation and according to the research strategy, are

[1,2,5]: strokes or the combination of compressing versus

Current Opinion in Food Science 2016, 9:21–28 www.sciencedirect.com

An updating about artificial mastication Peyron and Woda 23

shearing constraints allow adapting to different food

establishment of links between food fragmentation and

structures or to harder or more difficult food stuffs to

initial food structure [13].

chew.

4) Subjects with moderate impairment of the anatomical

Several mastication robots or mechatronic devices have

or physiological conditions of the masticatory appara-

been conceived and designed to study biomechanics of the

tus can also succeed in making a viable bolus through a

masticatory process. Development of a series of mastica-

more demanding adaptation. Again, the adaptation

tion robots was carried out for quantitative and dynamic

relies on increasing the force, the number of tooth

assessment of mechanical stress applied to oral elements

strokes or the constraint modes.

during oral activity. The ‘Waseda Jaw (WJ)’ systems were

mostly developed to analyze the mechanical effects of

Different kinds of simulation/reproduction of mastication on jaw bones in terms of position, force,

masticatory function velocities and muscle controls [14,15]. A second example

Biomechanical knowledge-oriented simulation of a mechatronic chewing device is of particular relevance

Computer or computer-assisted models have often been since it can reproduce the entire suite of complex functions

elaborated to analyze the dynamics of biomechanical and movements involved during mastication, encompass-

aspects of the masticatory function for dental, medical ing most of oral applications [16,17]. The main objective of

and therapeutic objectives and for understanding biolog- this device was to propose a ‘chewing robot’ (Figure 2a)

ical systems. It participates in predicting jaw movements, able to reproduce a molar trajectory in actual dimensions



muscle activations, recruitment patterns and controls, [18,19,20 ]. Aside from the area of food science, dentistry

resulting forces, or movements at the temporomandibular and specialists in dental materials developed tools to

joint [6–12]. Recently, some digital investigations based evaluate fatigue, resistance, wear or behaviour of restor-

on the discrete element method were conducted on the ative pieces under mechanical testing as close as possible to

food breakdown pathways during oral processing and the in vivo oral conditions [21–23].

Figure 2

Motor & control unit Crank

Ground link Coupler Link 6 Follower Link 5 Handle for the adjustable sagittal plane Quick teeth Handle for the attachment adjustable ground mechanism Mandible molar Shock absorber Food retention mechanism

Maxilla molar Maxilla molar repositioning Handle for the table adjustable maxilla

Current Opinion in Food Science

The ‘chewing robot’.

Reproduced with authorization from [30].

www.sciencedirect.com Current Opinion in Food Science 2016, 9:21–28

24 Sensory science and consumer perception

Food-oriented or bolus-oriented simulation of saliva of known flow and composition. These needs

The first attempts to mimic jaw movement with a main induced specific requirements that were very challenging

interest towards the food sample were equipment roughly for the conception of a masticatory device. It led to debat-

designed to activate the upper jaw against the food able choices; for example, in terms of food disruption

sample for measuring mechanical properties of food tex- modalities, the volume of the artificial mouth, or the dura-

ture or equipped for example with a piston presenting a tion of masticatory sequence, to name a few. The ‘artificial

cuspal angulation reflecting angles observed in the mouth mouth’ developed by Salles and collaborators (Figure 2b) is

[24,25]. This kind of machine, considered as providing probably the most successful apparatus for measuring aro-

objective methods for food evaluation, generally dis- ma release during chewing since it encompasses more

played significant correlation between sensory perception physiological purposes than others [36,38]. The apparatus

and mechanical measurement. Similarly, food science produces food breakdown due to two opposite tooth arches

researchers tried to improve the first basic devices devel- actuated in both vertical and horizontal/angular motions.

oped to describe food texture [26]. For example, the Volatile retention is completed with a gas introduced into

experimental ‘crush chamber’ was designed to include the system, allowing air sampling in synchronization with

evaluation of acoustic, tactile and olfactory stimuli during mastication events, as sniffing does in vivo. Food break-

crispbread mastication [27]; the ‘BITE Master II’ was down has only been ‘validated’ against peanut particle size

elaborated to study the perception of cheese hardness observed in vivo in a very few number of subjects [36].

during the very first chew [28], and an ‘in vitro mouth

2

model’ was developed for the determination of salt release The ‘AM apparatus’ (Figure 3b) is the unique mastication

from the food matrix [29]. The ‘chewing robot’ (Figure 2a) machine focusing on the food bolus as the result of masti-

was first developed to reproduce the mechanics of the cation while introducing most of the actual biomechanical

2

chewing process but could also be proposed in the future to masticatory features [39,40]. The AM apparatus thus

give a quantitative analysis of mechanical disruption allow- permits simulation of mastication in various oral contexts

ing texture analysis of a food sample in nutritional ques- and provides a complete food bolus recovery after masti-

tionings [30]. In addition, some simple instrumentation cation for further analysis. It produces a food bolus with

was developed for semi-solid food issues [31]. properties similar to those of a bolus produced by in vivo

mastication in numerous subjects ([41] — Figure 1). This

Since it leads to perception of flavour, the release of volatile kind of device can also be successively employed to

aromatic compounds during food disruption is one of the investigate food science, physiological or nutrition fields

issues most studied using chewing simulation [32–37]. In such as nutrient bioaccessibility assessment or digestive

these different approaches, the liberation or retention of process follow-up in link with oral food transformation

volatile molecules was measured in relation to the presence ([42] — Figure 1).

Figure 3

(a) (b)

gas sampling mobil e strength masticatory saliva upper jaw (fixed) sensor disc injection

tongue

teeth

lower jaw (mobile) saliva slid e for

lower jaw pump liquid and ton gue collection

detail of the fixed actuatio n mastication chamber masticatory

disc

(to be opened for bolus collection) Current Opinion in Food Science

2

(a) The ‘artificial mouth’ (reproduced with authorization from [36]), and (b) the ‘AM masticator apparatus’ [39].

Current Opinion in Food Science 2016, 9:21–28 www.sciencedirect.com

An updating about artificial mastication Peyron and Woda 25

Biomechanical aspects of masticatory applied to the food samples difficult. It also renders

simulators difficult the recovery of the food particles that constitute

Depending on the reason for using them, the various the bolus. Consequently, it could probably cannot be

existing mastication simulators have differently set five used for other purposes than study of the dynamic release

key variables: teeth or equivalent, inside-mouth volume, of volatiles during mastication. Finally, the use of dental

saliva or equivalent, temperature control, and kinetic and arcades similar to the ‘real’ anatomy has not been shown

stress modalities of functioning. The most crude simula- to give better correlation between sensory and instru-

tion of tooth function is probably Mills’s ‘in vitro mouth mental hardness assessment than when food hardness is

model’ that only compresses a food sample under a flat measured by a classical compression test. It may also

piston to measure the salt released in the liquid medium introduce another source of variation by its inability to

[29]. Other developers equipped their apparatus with maintain the food particles between the teeth. This limit

teeth using either a complete human skull (‘Waseda was accounted for in the artificial mouth of Salles’s team

Jaw’, [14]), patient’s complete arcades (‘Bite MASTER by a tongue placed at the centre of the ring supporting the

II’, [28]), or series of molar teeth fixed on two opposite teeth, programmed to place food particles on the teeth.

ring-shaped cylinder) ‘artificial mouth’, [36]). The major This design, however, does not gather food particles in a

limit of this type of choice is that it under-estimates the bolus since particles are inevitably distributed over the

role of the central nervous system in taking advantage of full ring [36]. Despite these limits, this latter device

the complex anatomy of the tooth arches. The control of seems to be the most advanced for the study of aroma

masticatory movements and forces performed by the release during oral food breakdown. The systems

nervous system cannot be replaced and this renders equipped with cutting blades [37] or triangular-shaped

difficult the interpretation of what happens to the food elevations [27], cannot be considered to mimic mastica-

sample in term of mechanical stress and strain. The tory action due to the absence of a lot of components of

experimental mouth proposed by Salles et al., with teeth movement, of ‘tooth’ elements and no control of the stress

2

organized on a circle-shaped design (Figure 3a), misco- applied to the food sample. In the AM apparatus, tooth

pies the normal human tooth contacts and offers more function is reproduced but not tooth anatomy. Tooth

contacts between teeth and food than in a human mouth, action is made by two opposite triangular forms whose

making the estimation of the forces and constraints active surfaces are similar to the sum of the molar and

Figure 4

coconut carrots 100 100 90 90 80 80 70 70 60 60 50 50 40 in vivo 40 30 in vitro (AM2) 30 in vivo 20 20 in vitro (AM2) 10 10 cumulative weight (%) 0 cumulative weight (%) 0 0.4 0.8 1 1.4 2 2.5 4 0.4 0.8 1 1.4 2 2.5 4 Sieve aperture (mm) Sieve aperture (mm)

pork meat 100 green olives 90 100 80 90 70 80 60 70 50 60 50 40 in vivo 40 30 in vitro (AM2) 30 in vivo 20 20 in vitro (AM2) cumulative weight (%) 10 10 0 cumulative weight (%) 0 0.4 1 1.4 2 2.5 4 6.3 7.1 0.4 0.8 1 1.4 2 2.5 4 Sieve aperture (mm) Sieve aperture(mm)

Current Opinion in Food Science

Comparison of particle size distributions obtained in food boluses collected at the end of mastication in vivo in volunteers with normal dentitions

2

or in vitro with the AM masticatory apparatus.

www.sciencedirect.com Current Opinion in Food Science 2016, 9:21–28

26 Sensory science and consumer perception

premolar surface areas involved when chewing a standard released from the food matrix. The bolus can be analyzed

bolus. These ‘tooth’ elements are actuated by translation for particle size distribution, a major characteristic of food

and rotational movements to ensure correct impact on disruption.

food and gathering of the particles before tooth confron-

tation [39]. Any chewing device used to provide food boluses has to

be validated against human mastication (Figure 4).

Three other key points are important in the develop- Such validation has not been conducted for many of

ment of a simulator. Saliva should be used. Ideally, its the proposed systems. This deficiency is striking in

composition, flow distribution along masticatory se- digestion studies, which are generally operated without

quence and total injected volume should mimic those a specific masticatory apparatus or with food particles

seen in the human mouth. The volume of the ‘mastica- coarsely ground or minced and mixed with saliva or

tory chamber’ should be similar to the volume of the enzyme during an uncontrolled or unjustified time, to

mouth and a possibility of controlling the oral tempera- obtain what must be considered as a fortuitous food

ture should exist. Saliva, volume and temperature items bolus [43]. Mishellany-Dutour et al. [41] validated

2

are fundamental for studying aroma or nutrient release the AM apparatus by comparing particle size distribu-

and food texture measurements. Not all apparatuses are tion and median particle size of an in vitro bolus with

equipped for these controls and this may affect data a bolus made in vivo by selected subjects with normal

interpretation. dentitions, a correct occlusion and a normal saliva

flow (Figures 4 and 5). Some bolus rheological proper-

The final items that should be considered are kinetic ties, hardness or cohesiveness, for example, are also

factors and constraint modalities of functioning. Various very informative of the suitability of the bolus to be

degrees of freedom have been chosen depending on the safely swallowed [5] and should also be used for in vivo/

main purpose for using the apparatus (aroma release, food in vitro validation purposes.

texture/bolus measurements, dental training, for exam-

ple). Obviously, complex mandibular movements adjust-

ed to the food being chewed cannot be completely

Figure 5

reproduced. Complete feedback control is always absent

although it has been sought while studying the first stroke

[28]. This requirement has been addressed differently by PEANUTS

choosing to reproduce or control the mechanical function, 4

jaw movements, imitation tooth anatomy and applied 2.49

3

forces [30,36,39], or by applying fracture propagation 2.34

knowledge to food matrix during disruption (tooth action

2

in mechanical terms) in order to select appropriate stress- 1.39 1.36 1.39 1.38

d50 (mm) strain conditions [39]. 1

0

10 20 26

Validation of mastication devices by food Number of masticatory cycles

bolus analysis

The food bolus is the main focus of interest in most topics

in vivo n=30 subjects

in food science research. Food bolus analysis is at the

CARROTS in vitro n=10 trials

crossroads between food structure, food formulation, food 9 6.74

8

perception, food oral processing and the further stages of 6.49

7

digestion. The ready-to-swallow bolus contains informa-

6

tion about the oral conditions of its formation. In addition, 5 3.88

3.67

4 it constitutes the vector for nutrients. For all these rea- 2.78 2.73

d50 (mm) d50 3

sons, a mastication simulator provides a valuable contri-

2

bution since it allows recovering the totality of the food 1

bolus at the end of the masticatory sequence. During 0

10 20 33

mastication, food sample is drastically disrupted to form a

Number of masticatory cycles

cohesive entity, which can be swallowed easily and with-

current opinion in food science

out risk of particle aspiration. As particles are formed, they

are mixed with saliva. During this process, the smaller the

Median particle size (d50 values) of food boluses collected after

food particles, the greater the surface contacts between

10 cycles, 20 cycles of at the end of the masticatory sequence, in vivo

food and saliva, favouring the access of salivary enzymes 2

in volunteers with normal dentitions and in vitro with the AM

to substrates. The ready-to-swallow final bolus is com- masticator apparatus.

posed of particles of various sizes and saliva or juice Reproduced with authorization from [40].

Current Opinion in Food Science 2016, 9:21–28 www.sciencedirect.com

An updating about artificial mastication Peyron and Woda 27

Conclusion This synthesis presents what results and physiological laws governing

mastication must be known before simulation of masticatory process.

In summary, when the major objective of simulation is to

2. Jalabert-Malbos ML, Mishellany-Dutour A, Woda A, Peyron MA:

reproduce the biomechanics of jaw movements and

Particle size distribution in the food bolus after mastication of

forces, mathematical models or robots are more appropri- natural foods. Food Qual Prefer 2007, 18:803-812.

ate. When the objective is to study the resulting food

3. Agrawal KR, Lucas PW, Prinz JF, Bruce IC: Mechanical

bolus, a device reproducing masticatory parameters and properties of foods responsible for resisting food breakdown

in the human mouth. Arch Oral Biol 1997, 42:1-9.

correct food disruption is a better option. The specifica-

4. Prinz JF, Lucas PW: An optimization model for mastication and

tions and technical limits for a simulation device depend

swallowing in mammals. Proc Royal Soc Lond Ser B Biochem

on the primary research purpose: aroma release, food

1997, 44:1715-1721.

texture assessment, production of a food bolus providing

5. Peyron MA, Gierczynski I, Hartmann C, Loret C, Dardevet D,

for subsequent digestion analyses, the impact of a change Martin N, Woda A: A likely role of physical bolus properties as

sensory inputs in swallowing triggering. PLoS ONE 2011,

in food formulation on bolus, biomechanical analysis of

6:e21167.

stress applied to oral elements and other factors are

6. Osborn JW, Baragar FA: Predicted pattern of human muscle

specific questions. Nevertheless, taking account of the

activity during clenching derived from a computer assisted

main laws governing the biomechanical mechanisms of model: symmetric vertical bite forces. J Biomech 1985,

18:599-612.

the masticatory processes and the dynamics of bolus

properties are always needed [26]. Knowing the features 7. Baragar FA, Osborn JW: Efficiency as a predictor of human jaw

design in the sagittal plane. J Biomech 1987, 20:447-457.

associated with a correct use of an in vivo bolus simulator

allows the preparation of a realistic bolus for subsequent 8. Koolstra JH, van Eijden TM, Weijs WA, Naeije M: A three-

dimensional mathematical model of the human masticatory

analyses (Figure 1). In addition, it favours interpretation

system predicting maximum possible bite forces. J Biomech

of the results. Obviously, all apparatuses must be validat- 1988, 21:563-576.

ed before use and this essential step can only be done by

9. Gallo LM, Airoldi GB, Airoldi RL, Palla S: Description of

comparing boluses obtained in vivo in a sufficient number mandibular finite helical axis pathways in asymptomatic

subjects. J Dent Res 1997, 76:704-713.

of individuals versus boluses obtained in vitro. Clearly, a

10. Slagter GEC, Otten E, van Eijden TMGJ, van Willingen JD:

mastication simulator cannot reproduce the large range of

Mathematical model of the human jaw system simulating

mastication strategies observable in human. The appara-

static biting and movements after unloading. J Neurophysiol

tus is therefore assumed to reflect average mastication. A 1997, 78:3222-3233.

well-designed apparatus can be employed to investigate 11. Koolstra JH, van Eijden TMGJ: A method to predict muscle

control in the kinematically and mechanically indeterminate

the kinetics of bolus formation in terms of particle size,

human masticatory system. J Biomech 2001, 34:1179-1188.

rheological behaviour, saliva impregnation, compound

12. Gal JA, Gallo LM, Palla S, Murray G, Klineberg I: Analysis of

release or biochemical modification, as a function of food

human mandibular mechanics based on screw theory and in

structure, food formulation and processing, or even to vivo data. J Biomech 2004, 37:1405-1412.

simulate several dental or ageing conditions (Figure 1).

13. Hedjazi L, Martin CL, Guessasma S, Della Valle G, Dendievel R:

Mimicking oral steps using a kitchen food processor or a Experimental investigation and discrete simulation of

fragmentation in expanded breakfast cereals. Food Res Intern

static oral digestion apparatus, as is too often done in

2014, 55:28-36.

digestion studies, cannot produce a realistic bolus for a

14. Kato I, Takanisi A, Asari K, Tani T: Development of artificial

majority of solid foods, especially if considering specific

mastication system. Construction of one degree of freedom

populations such as the elderly or infants, because it antagonistic muscle model WJ-O. Anat Anz 1988, 165:197-203.

cannot produce a particle size distribution and cannot

15. Usui T, Maki K, Toki Y, Shibasaki Y, Takanobu H, Takanishi A,

reproduce the dynamics of bolus formation. If these Hatcher D, Miller A: Measurement of mechanical strain on

mandibular surface with mastication robot: influence of

complex oral steps are overlooked, data obtained from

muscle loading direction and magnitude. Orthod Craniofac Res

bolus analyses in digestion studies or for texture assess- 2003, 6(Suppl. 1):163-167.

ment during the whole masticatory sequence could be

16. Xu WL, Bronlund JE, Kieser JA: Choosing new ways to chew.

biased. IEEE Robot Autom Mag 2005, June:90-98.

17. Pap JS, Xu WL, Bronlund J: A robotic human masticatory

Acknowledgements system: kinematics simulations. Int J Intell Syst Technol Appl

2005, 1:3-17.

The authors want to thank O. Franc¸ois (University of Auvergne) for his

2

involvement in the development of the AM apparatus and in preparing the

18. Daumas B, Xu WL, Bronlund J: Jaw mechanism modelling and

figures; and P. Riordan for English revision. simulation. Mech Mach Theory 2005, 40:821-833.

19. Xu WL, Bronlund JE, Potgieter J, Foster KD, Ro¨ hrle O, Pullan AJ,

References and recommended reading Kieser JA: Review of the human masticatory system and

Papers of particular interest, published within the period of review, masticatory robotics. Mech Mach Theory 2008, 43:1353-1375.

have been highlighted as:

20. Xu W, Bronlund JE: Mastication Robots. Biological Inspiration to

 Implementation. Studies in Computational Intelligence. Berlin:

 of special interest

Springer; 2010.

 of outstanding interest

This book depicts how the use of a robotic simulation of mastication

allows the study of biomechanics of the masticatory system.

1. Woda A, Foster K, Mishellany A, Peyron MA: Adaptation of

 healthy mastication to factors pertaining to the individual or to 21. Conserva E, Menini M, Tealdo T, Bevilacqua M, Pera F, Ravera G,

the food. Physiol Behav 2006, 89:28-35. Pera P: Robotic chewing simulator for dental materials testing

www.sciencedirect.com Current Opinion in Food Science 2016, 9:21–28

28 Sensory science and consumer perception

on a sensor-equipped implant setup. Int J Prosthodont 2008, (Phaseolus vulgaris) influenced by composition and volume of

21:501-508. artificial saliva. Z Lebensm Unters Forsch 1996, 203:1-6.

22. Raabe D, Alemzadeh K, Harrison AL, Ireland AJ: The chewing 34. Deibler KD, Lavin EH, Linforth RS, Taylor AJ, Acree TE:

robot: a new biologically-inspired way to evaluate dental Verification of a mouth simulator by in vivo measurements.

restorative materials. Conf Proc IEEE Eng Med Biol Soc. J Agric Food Chem 2001, 49:1388-1393.

2009:6050-6053.

35. van Ruth SM, Buhr K: Influence of mastication rate on dynamic

23. Steiner M, Mitsias ME, Ludwig K, Kern M: In vitro evaluation of a flavor release analysed by combined model mouth/proton

mechanical testing chewing simulator. Dent Mater 2009, transfer reaction-mass spectrometry. Int J Mass Spectrom

25:494-499. 2004, 239:187-192.

24. Friedman HH, Whitney JE, Szczesniak AS: The texturometer — a 36. Salles C, Tarrega A, Mielle P, Maratray J, Gorria P, Liaboeuf J,

new instrument for objective texture measurement. J Food Sci Liodenot JJ: Development of a chewing simulator for food

1963, 28:390-403. breakdown and the analysis of in vitro flavor compound

release in a mouth environment. J Food Eng 2007, 82:189-198.

25. Wang JS, Stohler CS: Textural properties of food used in

studies of mastication. J Dent Res 1990, 69:1546-1550. 37. Arvisenet G, Billy L, Poinot P, Vigneau E, Bertrand D, Prost C:

Effect of apple particle state on the release of volatile

26. Morell P, Hernando I, Fiszman SM: Understanding the relevance compounds in a new artificial mouth device. J Agric Food Chem

of in-mouth food processing. A review of in vitro techniques. 2008, 56:3245-3253.

Trends Food Sci Technol 2014, 35:18-31.

38. Mielle P, Tarrega A, Se´ mon E, Maratray j, Gorria P, Liodenot JJ,

27. Winquist F, Wide P, Eklo¨ v HC, Lundstro¨ m I: Crisbread quality Liaboeuf J, Andrejewski JL, Salles C: From human to artificial

evaluation based on fusion of information from the sensor mouth, from basic to results. Sens Actuators B 2010,

analogies to the human olfactory, auditory and tactile senses. 146:440-445.

J Food Process Eng 1999, 22:337-358.

39. Woda A, Mishellany-Dutour A, Batier L, Franc¸ois O, Meunier JP,

28. Meullenet JF, Gandhapuneni RK: Development of the BITE Reynaud B, Alric M, Peyron MA: Development and validation of a

Master II and its application to the study of cheese hardness. mastication simulator. J Biomech 2010, 43:1667-1673.

Physiol Behav 2006, 89:39-43. 2

40. AM , Artificial Mastication advanced Machine website; URL:

29. Mills T, Spyropoulos F, Norton IT, Bakalis S: Development of an http://masticateur.u-clermont1.fr/.

in-vitro mouth model to quantify salt release from gels.

41. Mishellany-Dutour A, Peyron MA, Croze J, Franc¸ois O, Hartman C,

Food Hydrocoll 2011, 25:107-113.

Alric M, Woda A: Comparison of food boluses prepared in vivo

30. Xu WL, Lewis D, Bronlund JE, Morgenstern MP: Mechanism, and in vitro by the mastication simulator: AM2. Food Qual

design and motion control of a linkage chewing device for Prefer 2011, 22:326-331.

food evaluation. Mech Mach Theory 2008, 43:376-389.

42. Hennequin A, Peyron MA, Ferreira C, Aubry L, Sante´ -Lhoutellier V:

31. Prinz JF, Janssen AM, de Wijk RA: In vitro simulation of the Addition of fibers in Frankfurters modifies the ready-to-

oral processing of semi-liquid foods. Food Hydrocoll 2007, swallow food bolus properties and oral bioaccessibility of

21:397-401. nutrients after in vitro mastication. In Proceedings of 61st

International Congress of Meat Science & Technology (ICOMST);

32. Nassl K, Kropf F, Klostermeyer H: A method to mimic and to Clermont-Ferrand, France, August: 2015.

study the release of flavour compounds from chewed food.

Z Lebensm Unters Forsch 1995, 201:62-68. 43. Minekus M, Alminger M, Alvito P, Ballance S, Bohn T, Bourlieu C,

Carrie` re F, Boutrou R, Corredig M, Dupont D et al.: A standardised

33. van Ruth SM, Roozen JP, Nahon DF, Cozijnsen JL, static in vitro digestion method suitable for food — an

Posthumus MA: Flavour release from rehydrated french beans international consensus. Food Funct 2014, 5:1113-1124.

Current Opinion in Food Science 2016, 9:21–28 www.sciencedirect.com